1: \begin{abstract}

2: 	

3: 	In this paper, we focus on the decentralized optimization problem over the Stiefel manifold,

4: 	which is defined on a connected network of $d$ agents.

5: 	The objective is an average of $d$ local functions,

6: 	and each function is privately held by an agent and encodes its data.

7: 	The agents can only communicate with their neighbors 

8: 	in a collaborative effort to solve this problem.

9: 	In existing methods, 

10: 	multiple rounds of communications are required to guarantee the convergence,

11: 	giving rise to high communication costs.

12: 	In contrast, this paper proposes a decentralized algorithm, called DESTINY,

13: 	which only invokes a single round of communications per iteration.

14: 	DESTINY combines gradient tracking techniques 

15: 	with a novel approximate augmented Lagrangian function.

16: 	The global convergence to stationary points is rigorously established.

17: 	Comprehensive numerical experiments demonstrate that DESTINY

18: 	has a strong potential to deliver a cutting-edge performance

19: 	in solving a variety of testing problems. 

20: 	

21: \end{abstract}

22: